»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend! | tinyurl.com/p6contest
Set by moritz_ on 28 December 2010.
masak anyway, now we're one character closer to cracking your IRC password. and you're not nearly paranoid enough. :P 00:01
colomon yes, but asking you to type a password and then reliably changing the focus to a public communications channel halfway through is a pretty massive flaw... 00:02
luckily I only use this password for IRC.
masak duly noted.
sjohnson colomon: what client are you using? 00:03
00:03 Limbic_Region joined
colomon Colloquy 00:03
sjohnson ahh, for mac
yeah, there's a lot of dumb UI designs in all sorts of software... i have my own share of rants too :)
anything that steals focus actually is a no-no in my book
00:04 rgrau left 00:05 rgrau_ joined
colomon rakudo: say (say "Found $_!" if $_ %% 10 for 11..12).perl 00:05
p6eval rakudo 2666b6: OUTPUT«()␤»
masak Emacs practically doesn't steal focus. 00:06
one of the nice things about it.
00:10 rgrau_ left
masak wow, I should sleep but the Yapsi refactor has taken off now, and it's hard to pull away and switch off. 00:11
I no longer have any doubts that introducing the FUTURE as an extra "layer" was the right decision. 00:12
00:14 ab5tract left 00:15 masak left, plobsing joined 00:17 masak joined, masak left
flussence do any perl5/6 test tools have anything for this? Sounds like a good idea: www.reddit.com/r/programming/commen...?context=2 00:49
01:11 cosimo left 01:15 drbean joined 01:29 cafesofie joined 01:36 felliott left 01:37 cafesofie left 01:41 felliott joined 01:43 felliott_ joined, felliott left, felliott_ is now known as felliott 01:46 felliott left 01:55 cafesofie joined 02:01 noganex joined 02:02 thundergnat joined
thundergnat rakudo: my ($a,$b); $a = $a / 4; $b /= 4; say "$a, $b"; 02:03
p6eval rakudo 2666b6: OUTPUT«Use of uninitialized value in numeric context in <anon> at line 22:/tmp/opsFIrtykw␤Method 'Bridge' not found for invocant of class 'Failure'␤ in 'infix:</>' at line 3729:CORE.setting␤ in main program body at line 1␤»
thundergnat my ($a,$b); $a = $a * 4; $b *= 4; say "$a, $b";
rakudo: my ($a,$b); $a = $a * 4; $b *= 4; say "$a, $b"; 02:04
p6eval rakudo 2666b6: OUTPUT«Use of uninitialized value in numeric context in <anon> at line 22:/tmp/ws1ajlzZFA␤0, 4␤»
02:04 noganex_ left
thundergnat The fact that unintialized $a *= 4 is equal to 4 is just bizarre... and it's spec too. 02:06
IMO that doesn't just _violate_ the priciple of least surprise, it violates it, beats it up, stabs it a few times and leaves it for dead in a drainage ditch.
colomon heh 02:09
I have to admit, I'm starting to long for a "use really-strict" which doesn't let you use unintialized variables or automatically convert arrays to scalars. 02:11
did you see moritz_'s mess today? 1/6 [+] @list
he wanted 1/6 * [+] @list 02:12
without the *, it magically becomes 1/6 + @list.elems
thundergnat Yeah, saw it in back log. 02:13
Makes sense after you think about it, but not real obvious.
colomon it makes perfect sense, but why would anyone want it to work that way? 02:14
02:32 Vlavv joined 02:33 fhelmberger joined 02:35 Chillance joined 02:42 thundergnat left 02:43 whiteknight left 02:49 cdarroch left 02:53 felliott joined
sorear good * #perl6 03:04
phenny sorear: 02 Feb 22:19Z <jnthn> tell sorear answers in gist.github.com/808572
sorear :p
jnthn: 6model seems to assume that if a type check "cache" is present, it is exhaustive 03:07
colomon: +1 to warn on "useless use of grouping brackets" 03:11
03:11 Su-Shee left, Su-Shee joined
colomon sorear: sounds like a fine plan, if it can reasonably be done. 03:14
sorear phenny: tell jnthn 6model seems to assume that if a type check "cache" is present, it is exhaustive 03:20
phenny sorear: I'll pass that on when jnthn is around.
sorear phenny: tell jnthn Where is the code in nqp-nom that handles $!str_native_attr := "foo"; unboxing? Is "foo" a parrot;Str or a P6opaque something? 03:21
phenny sorear: I'll pass that on when jnthn is around.
sorear phenny: tell jnthn What src/* directories are used in nqp-nom and what tests are passed? 03:22
phenny sorear: I'll pass that on when jnthn is around.
03:24 cosimo joined 03:28 mikehh left 03:30 Limbic_Region left 03:38 uvtc joined
sorear nqpnom: say(1) 03:38
nqpclr: say(1)
nqpnet: say(1)
nom: say(1) 03:39
p6eval nqpnet: OUTPUT«(timeout)»
sorear 6model: say(1)
03:39 stifynsemons joined
uvtc Hi. I'd like to install Rakudo Star into its own /usr/local/rakudo directory. During the "perl Configure.pl" step, should I use --prefix=/usr/local/rakudo, or --get-parrot-prefix=/usr/local/rakudo? 03:39
sorear --gen-parrot-prefix 03:40
uvtc sorear: Thanks, sorear. What does --prefix do?
03:40 karupanerura joined
sorear oh wait, you said rakudo _star_ 03:40
I don't know what you should use then
uvtc Yes.
Hm. 03:41
I'd installed Rakudo from source in the past, and used --gen-parrot-prefix. This is my first Rakudo Star install.
03:41 patspam joined
uvtc The README says, "The "--prefix=" option can be provided to Configure.pl to change the location of the install directory.", but I'm not sure if they're talking about Parrot or Rakudo. 03:42
sorear Rakudo Star is supposed to hide Parrot 03:43
uvtc Right.
Sorry, s/--get-parrot-prefix/--gen-parrot-prefix/ 03:46
And I'm also specifying --gen-parrot, of course. 03:47
03:51 felliott left 03:54 felliott joined
uvtc I don't see "gen-parrot-prefix" anywhere in the Configure.pl script. Going with just plain "--prefix"... 03:59
04:00 satyavvd joined 04:14 plobsing_ joined 04:17 plobsing left 04:23 Su-Shee_ joined 04:24 jarfhy left 04:26 Su-Shee left
uvtc sorear: Seems to work fine. Thanks. :) 04:27
dalek ecza: 0451061 | sorear++ | src/NieczaCompiler.pm6:
Force a message when compiling setting (colomon)
 04:29
colomon sorear++
sorear wait, there's more. 04:32
dalek ecza: 36d5b58 | sorear++ | src/niecza:
Warn about useless-looking [+]
 04:33
colomon nice. 04:36
sorear niecza: 1 [+] 2 04:41
p6eval niecza v2-1-gc6596ea: ( no output )
05:03 uvtc left
sorear niecza: class Boo { has $.x; method INVOKE($y) { $.x + $y } }; my $adder = Boo.new(x => 5); say $adder(7) 05:06
p6eval niecza v2-1-gc6596ea: OUTPUT«12␤»
05:07 patspam left
sorear wasn't expecting that to work. 05:10
05:11 Sarten-X joined
sorear phenny: tell moritz_ /chroot/home/p6eval/log/6model.log # Broken build! 05:11
phenny sorear: I'll pass that on when moritz_ is around.
sorear phenny: tell jnthn Under what circumstances does nqp-nom ever use P6str? 05:18
phenny sorear: I'll pass that on when jnthn is around.
dalek ecza: f3cce73 | sorear++ | docs/nam.pod:
Document more nam ops for pmurias
 05:23
sorear phenny: tell pmurias I finished your last docs-wanted list; let me know if any are insufficient
phenny sorear: I'll pass that on when pmurias is around.
sorear pmichaud: I don't suppose you can answer any of my design questions about 6model? 05:29
pmichaud sorear: probably not 05:30
05:31 alester left
pmichaud at least, not tonight. In a week or so I might be able to help out :) 05:31
05:31 envi joined
sorear What happens in a week? 05:32
pmichaud I'll have started working more closely with nqp-nom :)
05:37 jarfhy joined 05:39 Benabik joined
PerlJam pmichaud: What's the temperature in Plano? 05:40
s1n PerlJam: it's about 15F, 0-10F windchill, about 5 minutes north of pmichaud 05:43
(where i am)
PerlJam s1n: any snow? 05:44
s1n not anymore, got sleet and some snow on monday 05:45
it all froze and shutdown dfw
rolling blackouts and everything
triple digits for several months we can handle, sub-zero temps and it's crisis mode 05:46
sorear -> sleep 05:47
PerlJam We supposedly had rolling blackouts down here (Corpus Christi) but I didn't notice anything
s1n PerlJam: you're in corpus? going to txlf?
PerlJam what's txlf? 05:48
oh linux fest 05:49
no, I hadn't planned on it.
05:50 kaare_ joined 05:52 molaf joined 05:53 cafesofie left 05:57 lue joined 06:03 noganex_ joined 06:07 noganex left 06:08 noganex_ left, noganex joined, starcoder left 06:12 starcoder joined
Benabik 06:13
Benabik left, Benabik joined 06:14 molaf left
Benabik Not used to this client... Lemme see if I can dig up the old topic. 06:15
Took a bit of log searching, but this is what I found.. 06:20
Benabik Ack! 06:20
􏿽xBB􏿽xF6􏿽xAB Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend! 06:20
Benabik Now to figure out if I can remove that button from the toolbar. *grumble, grumble* 06:21
06:26 stifynsemons left 06:27 cotto joined 06:35 ggoebel left 06:36 ggoebel joined 06:38 cosimo left 06:42 karupanerura left 06:45 justatheory left 06:46 Mowah joined 07:01 wtw joined 07:46 cjk101010 joined 08:10 pnu joined 08:16 Su-Shee_ is now known as Su-Shee 08:57 abraxxa joined 09:32 masak joined
masak \o/ 09:32
tadzik o/
masak heads off for $work 09:37
09:39 dakkar joined 09:54 wamba joined
sjohnson $work eq $job 09:58
10:04 daxim joined 10:11 ab5tract joined 10:12 masak left
jnthn . 10:16
phenny jnthn: 03:20Z <sorear> tell jnthn 6model seems to assume that if a type check "cache" is present, it is exhaustive
jnthn: 03:21Z <sorear> tell jnthn Where is the code in nqp-nom that handles $!str_native_attr := "foo"; unboxing? Is "foo" a parrot;Str or a P6opaque something?
jnthn: 03:22Z <sorear> tell jnthn What src/* directories are used in nqp-nom and what tests are passed?
jnthn: 05:18Z <sorear> tell jnthn Under what circumstances does nqp-nom ever use P6str?
jnthn sorear: type check cache is presntly exhaustive. May or may not change that when I do subsets or other bits. May be a flag. Sometimes we want to reject ast too. 10:17
10:17 kensanata joined
jnthn er 10:17
phenny: tell sorear type check cache is presntly exhaustive. May or may not change that when I do subsets or other bits. May be a flag. Sometimes we want to reject ast too. 10:18
phenny jnthn: I'll pass that on when sorear is around.
jnthn phenny: tell sorear the thing that handles the boxing/unboxing is NYI, as I mentioend yesterday. P6str is used as the repr of str. In the case you mention "foo" should never get boxed, as it's a waste. 10:20
phenny jnthn: I'll pass that on when sorear is around.
jnthn phenny: tell sorear best thing on the native types is to wait a bit before copying it. I've some pieces still to work out the precise details of. 10:21
phenny jnthn: I'll pass that on when sorear is around.
moritz_ rakudo: my @a; say @a[^5].elems 10:28
phenny moritz_: 05:11Z <sorear> tell moritz_ /chroot/home/p6eval/log/6model.log # Broken build!
p6eval rakudo 2666b6: OUTPUT«5␤»
tadzik rakudo: my @a; say @a[^5].elems; say @a.perl
p6eval rakudo 2666b6: OUTPUT«5␤[]␤»
moritz_ that's certainly a bug 10:29
.perl and .elems don't agree
10:31 yarp joined
tadzik I think @a[^5] returns [ Any(), Any()...] 10:33
rakudo: my @a; @a.[^5].perl.say
p6eval rakudo 2666b6: OUTPUT«(Any, Any, Any, Any, Any)␤»
tadzik hence the behaviour
moritz_ rakudo: $_ = 0; my @a; $_++ for @a[^5]; .say 10:34
p6eval rakudo 2666b6: OUTPUT«Cannot modify readonly value␤ in '&infix:<=>' at line 1␤ in <anon> at line 7454:CORE.setting␤ in main program body at line 1␤»
tadzik maybe it should return an empty list
moritz_ d'oh, I thought I had used .perl on the slice
rakudo: my $c = 0; my @a; $c++ for @a[^5]; $c.say
p6eval rakudo 2666b6: OUTPUT«5␤»
tadzik karma c? 10:36
aloha c? has karma of 0.
tadzik hrpmh
karma c
aloha c has karma of 220.
moritz_ karma c++ 10:37
aloha c++ has karma of 0.
baest c++++ 10:38
tadzik poor C++, gives his father all the credit 10:39
xinming Just now, I try to read the synopsis again, which made me think, What make perl 6 still perl. though a much more powerful version of perl. :-) 10:42
sjohnson probably cause it was larry's walls idea, and a lot of perl guys are excited about perl 6 10:43
xinming sjohnson: Nope, Not because of lwall's idea.
moritz_ xinming: there are lots of discussions of the form "what makes perl perl?" on the internet; they also answer the question what makes Perl 6 perl
sjohnson xinming: if you're so sure it's not his idea, why are you even asking 10:44
xinming moritz_: What I can see is, sub subname { } construct.
I'll google a bit.
It seems, in perl 6 world, Everything can be changed.
tadzik You know, just because you can it doesn't mean you should 10:45
moritz_ you can also change everything in perl 5
so that part stayed the same :-)
tadzik Also, I don't think Perl 5 is much different in thes case, see perl5i and MooSex::Declare
moritz_ or source filters
xinming but perl is really different. 10:46
moritz_ which perl? different from what?
xinming I try to learn python, and even though I read the tutorials, I found doing things in perl way makes me more confortable.
moritz_: I don't know, that's why I am asking.
tadzik For it's more like a natural language, hence more natural to use once you get to know it. IMHO 10:47
sjohnson xinming: how do you know it's not larry walls idea? 10:48
moritz_ xinming: perlgeek.de/blog-en/perl-6/is-perl-....writeback 10:49
10:56 jedai left 10:58 cafesofie joined 11:12 icwiener joined 11:24 jedai joined 11:27 masak joined 11:37 [Coke] joined 11:42 satyavvd left 11:56 Vlavv left 12:11 Vlavv` joined, bluescreen joined 12:21 Vlavv` left 12:25 am0c joined
colomon just got Math::BigInt working on his linux box! o/ 12:26
moritz_ \o/ 12:27
colomon turns out that the GNU tools install shared libraries somewhere my linux doesn't look for them. sigh. 12:29
moritz_ add the dir to /etc/ld.so.conf, and run ldconfig 12:30
colomon yes, done, that's how I got it working. :)
where were you at 4am when I needed to know that. ;) 12:31
(probably right here, if I'd just thought to ask.)
moritz_ at 4am of "my" time: awake, but not on IRC :-)
colomon 4am my time, 3.5 hours before now 12:32
12:33 Vlavv` joined
moritz_ yep, I was online back then :-) 12:33
colomon shoot, if I'd just read the verbiage spat out by my "make install" I'd have learned the trick as well. sigh. hazards of programming in your sleep. 12:37
12:41 cotto left
dalek osystem: abee203 | (Solomon Foster)++ | projects.list:
Add Math::BigInt.
 12:47
colomon Well, that was an adventure! 12:49
masak \o/ 12:50
12:51 Vlavv` left 12:54 stifynsemons joined 12:58 stifynsemons left 13:00 coldhead left, stifynsemons joined 13:04 Vlavv` joined
takadonet morning all 13:04
colomon \o 13:06
masak morning, takadonet. 13:07
13:10 stifynsemons left 13:26 felliott left, felliott joined 13:27 MayDaniel joined
moritz_ today I learned that Mathematica has list assignment 13:31
it's a cool language. Sad there's only a proprietary implementation of it 13:32
masak why is that, by the way?
I mean, why hasn't someone made an open clone of it, like with Matlab?
moritz_ computer algebra systems are hard to write 13:33
there are open ones, but they all have their own syntax
and none of them are as good as Mathematica
13:34 cafesofie left
moritz_ I'd love to change that, but there's only so much you can do in life :-) 13:34
masak ah, priorities.
I wonder if there'll be a revolution in life extension in our lifetime. 13:35
moritz_ right. There a general purpose programming language I'm involved with, which eats up most of my -Ofun programming time
masak that'd change the landscape somewhat.
moritz_ lifetime extension alone isn't good enough
masak moritz_: when going to interviews now, I'm very grateful I've spend as much time as I have idling away with Perl 6... :) 13:36
moritz_ if I'm senile with 75 or 80, I can't write a CAS in the remaining 25 or 30 years, or however much I'd get :-)
masak: why?
masak right. senility will have to go as well.
moritz_ do they ask lots of stuff that p6 taught you?
masak yes, almost incidentally things tie back in with the p6 project. 13:37
mostly high-level stuff, like build systems and continuous integration.
moritz_ I thought you already had a new job... still doing interviews?
masak it's... complicated.
jnthn :)
masak there are... layers of consultancy...
jnthn Consultancy is like an onion. 13:38
masak it's smelly?
moritz_ :-)
jnthn I was thinking of the layers but... :P
moritz_ but can taste good if fried and served with the right other ingredients :-)
13:38 uvtc joined
masak that's what we serve: deep-fried consultancy :P 13:39
jnthn
.oO( Consulting bahji )
masak om nom nom
13:39 molaf joined
moritz_ masak: I think when next we meet in RL, you must share the complicated story in more details :-) 13:39
masak gladly.
flussence "Don't use mojolicious, we don't want cruft, use cgi instead" 13:40
flussence headdesk
_sri xD
flussence sometimes I really hate my $dayjob.
masak flussence: have you considered getting another one?
one that doesn't think that a web app framework is cruft? 13:41
takadonet flussence: which country are you from?
13:41 drbean left
flussence UK, but I'm not on the mainland which makes finding good places hard :( 13:41
13:44 MayDaniel left
flussence oh well, I'm familiar with writing crap web code already, I used php for 8 years. 13:45
masak PHP isn't just some truck that you can dump things on. it's a dumb pipe between the database and the web frontend. 13:46
13:46 felipe joined 13:48 patspam joined 13:50 plainhao joined 13:58 wamba left
uvtc Hi. Where can I find the most recent Emacs and Vim support files for Perl 6? 14:02
masak github.com/petdance/vim-perl 14:03
for Emacs, I'd recommend cperl-mode. it's not perfect, but it's very robust so it handles even Perl 6 well. 14:04
14:04 bluescreen left
uvtc Thanks masak. Do you know if anyone's working on a separate Perl 6 Emacs mode? 14:04
masak uvtc: for more tips, see perl6.org/whatever/
14:05 stifynsemons joined, timbunce joined
masak uvtc: hm, no, not that I can recall. 14:05
daxim upgrade to vim 7.3 to include the most recent vim-perl repo
uvtc masak: Missed that. Thanks. :)
masak uvtc: the underlying problem is that to parse Perl 6 really well, you have to *be* STD, or contain it.
uvtc: and STD is written in Perl 6. you see the circularity there.
uvtc I see. At the moment, just having syntax highlighting for things like comments and strings would be enough. 14:07
masak uvtc: example: string parsing. simple deal, right? nope -- interpolations basically pull in the parser for the rest of the language. code blocks do the same. so it's no longer a simple matter of "find the first unescaped quote that looks like the start quote".
so even string parsing is non-trivial. 14:08
uvtc Ah.
masak comments should be slightly easier, if you discount Pod.
moritz_ Pod isn't too hard to parse
uvtc What does Perl 6 use for comments besides "#"?
masak it is an open question how far/well you can actually fake all this without being STD.
uvtc: see S26.
uvtc masak: thanks. 14:09
masak github.com/perl6/specs/raw/master/...tation.pod 14:10
felliott morning, perl6! 14:11
masak felliott! \o/
felliott I think i found a bogus test, and I'd like to confirm with you.
masak!
masak go right ahead. 14:12
14:12 mtk0 joined
masak had the urge the other day to do a writeup: "S26 and what we need to do about it" 14:12
felliott rakudo: my $x = 42; @p = $x += 6, 7; @p[1].say;
p6eval rakudo 2666b6: OUTPUT«===SORRY!===␤Symbol '@p' not predeclared in <anonymous> (/tmp/MVNYt7d7Dn:22)␤»
felliott rakudo: my $x = 42; my @p = $x += 6, 7; @p[1].say;
p6eval rakudo 2666b6: OUTPUT«7␤»
felliott That's correct, right? 14:13
moritz_ += is always item assignment
so yes, I think it's correct
felliott The test says it should be 47, but I think that was a typo.
I don't even know where 47 would come from
Cool, thanks! 14:14
moritz_ assign.t is a mess.
felliott indeed
moritz_ one thing that could be done about it is splitting all list-vs-item-assignment tests into a separate file 14:15
then it would still be messy, but the mess would be mostly factored out
dalek ast: ba92630 | (Fitz Elliott)++ | S03- (2 files):
[reduce.t,short-circuit.t] unfudge infix:[orelse] tests
 14:17
ast: 032e94e | (Fitz Elliott)++ | S03-operators/assign.t:
[assign.t] fix and unfudge two broken tests
14:17 bluescreen joined
14:17 wooden joined, wooden left, wooden joined
felliott derp. 14:17
I forgot that I had already unfudge the orelse tests.
I better commit my orelse fix to rakudo.
moritz_ first step: don't panic :-) 14:18
masak don't panic, or else :P
moritz_ :-)
uvtc masak: re. S26, any idea why it's languishing? My impression is that much Perl 6 documentation is still being written in Perl 5 POD.
sjn 0th step: don't be on fire
uvtc masak: perhaps "languishing" is the wrong word. 14:20
masak I think "languishing" captures it pretty well. 14:22
14:22 patspam left
masak there have been efforts to implement S26. some of them quite successful. none of them have ended up in Rakudo or some other compiler, for various reasons. 14:22
I plan to give it a go at some point. 14:23
one of the main problems has been pure parsing speed. mberends++ ran into some of those limits.
TheDamian's module on CPAN is fairly fast, but it's in Perl 5, so that "doesn't count". 14:24
moritz_ and it's not up to date either
masak that, too.
to get the nice two-way integration required by S26, it has to be Perl 6 code in the compiler.
uvtc Hm. I wouldn't think parsing speed would be an issue for rendering docs.
masak welcome to the weird world of Perl 6. 14:25
uvtc That is, I don't really mind waiting a moment after typing "perldoc foo".
masak the future is already here, but it's coming to us through a greasy layer of s.l.o.w.
uvtc: man, you have no idea.
:) 14:26
s/a moment/1e2 moments/
speed: "hey, I can read the text as it scrolls by!"
uvtc Hahaha. So, it's like hearing everything through a Cap't Kirk translator: "must ... render ... these ... docs into ... a man page!" 14:27
masak *lol*
flussence I get that with perl5 perldoc, but you've heard enough work horror stories from me for today :)
moritz_ somebody wants to apply github.com/rakudo/rakudo/pull/10 ? the patch looks fine to me, but I currently don't have access to a machine where I could commit it 14:28
14:30 mtk0 left
uvtc Does the `perl6` repl have any sort of built-in help system? 14:30
moritz_ nope 14:31
uvtc moritz_: thanks
masak at some point, maybe there'll be a way to get the Pod for a certain core method. it'd definitely be nice. but we're not there yet. 14:35
moritz_ wonders about good ways to document multis and their individual candidates 14:36
masak a very good question. one S26 should address, but (in my recollection) doesn't.
we need more application developers to stress-test S26! :) 14:37
uvtc In the past, I've used my own made-up convention for marking API docs. Ex., starting the first line of the comment with "#/" or something. 14:41
masak uvtc: we're still at the point where blue-skying of that kind -- especially presenting whole proposals of "nice-to-have" features, tooling, etc -- is very much welcome and appreciated. 14:43
I, for one, would like to see a tool that took a .pm file (or a set of .pm files) as input, and output an .html file (or set of 'em) with impeccable API documentation. 14:44
dalek ast: 6f2f05b | (Fitz Elliott)++ | S03-operators/short-circuit.t:
[short-circuit.t] refudge orelse until patch is accepted
ast: f890121 | (Fitz Elliott)++ | S03-metaops/reduce.t:
[reduce.t] add new fudged tests for orelse; refudge existing tests
felliott *whew*
masak is always happy to see roast activity 14:47
moritz_ too
14:48 yarp left
felliott even when it's just undoing previous activity? :) 14:48
masak pfeh, details :P 14:49
felliott (patching while trying to get ready for work)-- 14:50
14:51 jerome left
uvtc masak: would be interesting to hear a blog post on the docs and S26. I don't know if S26 is difficult to implement or not -- perhaps someone should declare and plan a worldwide Perl 6 doc day to try. :) 14:56
14:56 mtk0 joined 14:57 jerome joined
masak uvtc: I've once made a Pod grant application draft. hold on... 14:58
masak needs a way to search his gists quickly :)
14:58 felliott left, mtk0 left
masak uvtc: gist.github.com/500732 14:59
14:59 mtk0 joined, mtk left 15:00 mtk0 left, shortcircuit left, shortcircuit joined
masak uvtc: I would like to apply to TPF for that grant. but I'm stuck on week 19 out of 20 on an old grant, and I'd very much like to get unstuck with that first, and wrap it up in all ways possible before starting a new one. 15:00
15:00 mtk joined
masak time and enthusiasm seems to be the limiting factors for doing that. 15:00
s/seems/seem/ 15:01
uvtc masak: what's the old grant that you're wrapping up? 15:02
masak: would be great to see that new grant proposal go through. 15:03
15:08 Mowah left
moritz_ the old grant was Web.pm 15:08
15:09 hudnix left
masak yes. I'm even partway through week 20, with no real blockers in sight. :/ 15:09
[Coke] masak: do you mean the normal TPF grant process, or hague? 15:10
masak the Pod one would probably be a Hague grant, true. 15:12
the Web.pm one is a TPF grant.
uvtc Ah, interesting. Web.pm. I remember thinking "what about Web.pm" when I saw CGI-App on modules.perl6.org. Then just recently I saw tadzik blog about Bailador.
moritz_ Hague grant requires some non-zero degree of compiler hacking
masak moritz_: well, see the grant application draft above. 15:13
moritz_ uvtc: I started porting CGI::App to Perl 6. It's runnable, but lacks things like integration with cookie module etc.
masak uvtc: to a certain extent, other modules have flowed in where Web.pm meant to be but currently isn't. :/
moritz_ masak: yes, I was explaining why Web.pm was not Hague
masak ah, ok.
15:14 cjk101010 left
moritz_ will wait with his grant proposal after tuits have become more predictible 15:16
15:17 cotto joined
masak the exceptions grant? looking forward to that. 15:18
moritz_ yes, that one 15:19
exceptions are too fiddly (and not -Ofun enough) for me to attack it without a grant 15:20
masak :)
moritz_ but still enough -Ofun to do it with a grant, and worth doing
masak nod
same with S26. it can be made fun with the external quid-pro-quo challenge a grant provides. 15:21
& 15:24
15:29 wtw left 15:30 stifynsemons left, Sarten-X left
xinming I was thinking, Is the word package just be a word which is used to mean the file is perl 5 package? 15:33
15:34 Trashlord left
xinming If we really need a perl 6 package, We need to use module. 15:34
Or, I missed something between package vs module.
15:35 Trashlord joined 15:36 timbunce left 15:39 ggoebel left, orafu left 15:41 Sarten-X joined 15:44 Trashlord left 15:45 Trashlord joined
xinming How do we add additional search path in rakudo? 15:48
The old use lib 'lib'; doesn't work it seems.
15:48 rbuels joined
moritz_ use the PERL6LIB env variable 15:49
or BEGIN { @*INC.push: 'path/you/want' }
15:49 MayDaniel joined
xinming thanks. 15:50
Now, wish to learn perl 6 again. :-)
15:56 MayDaniel left 15:57 Trashlord left, Trashlord joined 16:00 am0c left 16:01 patspam joined, patspam left, patspam joined 16:07 justatheory joined 16:08 ab5tract left 16:09 felliott joined 16:10 spq joined 16:14 Patterner left 16:16 Psyche^ joined, Psyche^ is now known as Patterner
uvtc I'm having a problem installing neutro. When it goes to build perl6-File-Tools, it fails with a segfault. 16:49
This is using the most recent Rakudo Star. On Ubuntu 10.04. 16:50
16:50 risou joined
uvtc I've got R* installed with --prefix /usr/local/rakudo. 16:50
dalek ast: 0484877 | (Fitz Elliott)++ | S03-operators/assign.t:
[assign.t] add fudged tests for orelse
 16:51
uvtc It's a fresh install, with no other modules installed yet.
Anyone else run into this? 16:52
pmichaud uvtc: I haven't heard of it. Maybe file a ticket in RT or on github in the star repo 16:58
or with neutro
uvtc Thanks, Pm. 16:59
jnthn evening :) 17:02
pmichaud: How's the nqp move going? 17:04
17:04 kensanata left 17:17 daxim left
moritz_ btw nom doesn't build when an installed version of itself is available 17:22
moritz_ opens a ticket 17:23
17:24 abraxxa left 17:25 risou left
dalek p-cl-parser: 0427c30 | moritz++ | do.pl:
rename classes to live in HLL::
 17:26
17:27 kjeldahl joined
tadzik colomon: I don't think you got dependencies in META.info right 17:37
17:37 stifynsemons joined, stifynsemons left 17:41 cdarroch joined, cdarroch left, cdarroch joined 17:43 pmurias joined
pmurias masak: re only Perl 6 can parse Perl 6 causing problems for emacs, compiling STD to emacs lisp should solve the problem ;) 17:44
phenny pmurias: 05:23Z <sorear> tell pmurias I finished your last docs-wanted list; let me know if any are insufficient
pmurias tell sorear great, i'll try to implement that stuff soonish 17:45
17:51 dukeleto left, dukeleto joined
jnthn moritz_: ouch. 17:52
moritz_: Doesn't make much sense to me. :/
szbalint anyone coming to FOSDEM?
17:54 cafesofie joined 17:55 MayDaniel joined
pmichaud jnthn: haven't done the nqp move yet -- will probably try to do it today 17:57
17:57 rgrau joined
pmichaud Here we are, kids have snow day #3 home from school today. And the weather forecast for tonight/tomorrow morning is snow. :-/ 17:58
jnthn Ouch! 17:59
Snow...wow. I didn't associate that with your chunk of the world. :)
pmichaud it snows here, but never usually enough to cause this many closures
18:00 dakkar left
pmichaud and what we had earlier in the week was more ice than snow. snow would've melted off by now. the roads here have solid sheets of ice -- more like ice skating rinks than anything else 18:00
jnthn ugh
That's nasty.
pmichaud this was last year's big snowstorm (record snowfall for this area): picasaweb.google.com/lh/photo/DIN_...directlink 18:01
but by the next day it had mostly melted away
szbalint: I wish I was going to FOSDEM this year. I had a great time there at FOSDEM 2008. 18:02
that's a really good conference 18:03
szbalint last year was nice too :)
Perl presence is increasing aswell, this year we have a devroom apparently
szabgab++ 18:04
18:04 bluescreen left 18:05 Benabik left
jnthn Time to find some noms. :) 18:05
18:05 benabik joined 18:11 benabik left, ssps joined, benabik joined
ssps the thing I hate most in Perl 5 is the need to decode and encode every textual string, that comes and goes out of program. is this issue solved in Perl 6? 18:12
moritz_ ssps: it's a fundamental problem in any programming system. Perl 6 "solves" it by assuming an autodetected Unicode encoding by default 18:13
and it fixes the problem of accidentally mixing byte strings and text strings by packing them into (incompatiable) types 18:14
ssps encoding of what? of input, of stored data or of output?
moritz_ well, $*IN, $*OUT and open() default to that 18:15
ssps and which form this Unicode data really takes? 16-bits? UTF-8? 18:16
moritz_ ssps: see S32/IO
18:17 bluescreen joined
moritz_ perlcabal.org/syn/S32/IO.html 18:17
ssps oh, thank you.
18:17 MayDaniel left
pmichaud afk, too many things happening around the house 18:19
bbl
tadzik hmm,uri is NYI? Seems to be a LHF
moritz_ don't believe anything about the roles in S32/IO 18:20
tadzik It's not about the Roles 18:21
moritz_ ok :-)
tadzik I don't know how to link to the line in specs
oh, look: perlcabal.org/syn/S32/IO.html#uri_ 18:22
moritz_ look, an URI! 18:23
tadzik “These can naturally be overridden or added to by other modules” – I don't think I get it 18:24
ssps perlcabal.org/syn/S32/IO.html#IO%3A...%3AEncoded
not a line, but a title. pretty readable.
thanks, all. bye. 18:27
18:27 ssps left
benabik tadzik: I think it's referring to the fact that if you "use Net::SSH", IO.uri might start understanding scp:// URLs. 18:27
tadzik: Or "HTTP::Proxy" might override http:// URIs to hit a proxy first. 18:28
tadzik so it would augment class IO and feed it with a multi uri($uri where { $uri ~~ /^^scp:/ }), aye? 18:34
benabik If I'm reading it right, thats the purpose of the %*PROTOCOLS variable. `%*PROTOCOLS<scp> = SSH::Scp` and IO.uri will call `SSH::Scp.uri($uri)` when it gets a scp:// 18:36
tadzik ih, I see 18:41
18:42 vmspb joined
tadzik I'll put that on my TODO 18:42
18:46 pmurias left 18:58 envi left 19:16 Mowah joined 19:28 fhelmberger left
araujo now is single he wants to get a gf again 19:31
19:35 coldhead joined 19:36 lichtkind joined 19:38 stifynsemons joined 19:45 vmspb left 19:52 Sarten-X left
[Coke] Pretty sure this is the wrong channel forthat. 19:53
19:57 MayDaniel joined
colomon tadzik: can you be a little more specific? 19:58
20:00 Sarten-X joined 20:02 acrussell joined, acrussell left
tadzik colomon: sure thing. foo => [bar] is not a valid JSON, should be foo => ['bar']. Also, zavolaj is named NativeCall 20:11
20:11 Layla_91 joined
colomon fixed and pushed. tadzik++ 20:13
tadzik colomon: do you have a commit bit to my ecosystem fork?
Layla_91: hey!
colomon tadzik: not as far as I know?
tadzik colomon: you have 20:14
colomon okay.
Layla_91 tadzik: helo! =)
colomon tadzik: what makes you ask?
tadzik colomon: so you can add your module there 20:15
20:15 molaf left
colomon ah 20:15
tadzik if you want :) Otherwise I'll probably put it on my todo
colomon go ahead, though I might get around to it later if you haven't. :) 20:16
tadzik I'll TODO it then :)
20:22 vmspb joined 20:26 _jaldhar left
lichtkind cheers 20:27
ruoso was wishing javascript had coroutines...
Layla_91 tadzik: I need to download a file and keep getting disconnected every 20-30 minutes because of bad internet :( Do you know a good download manager that can resume if internet go down again?
flussence you could try wget if you don't mind command line 20:28
wget -c $url; repeat until done 20:29
Layla_91 flussence: but if the network connection goes down again, will wget resume or restart?
tadzik Layla_91: yeah, wget -c will continue
Layla_91 tadzik: wow.. I never noticed this option! hihi ^_^ 20:31
flussence you'll probably want --tries=0 --waitretry=5 too
benabik curl -C - -O $url does the same thing.
Just for more options. :-) 20:32
flussence yeah, but curl doesn't (apparently) have a way to say "retry indefinitely"... 20:33
mdxi put it in a loop until it returns success? 20:34
flussence --tries=0 is more OS-independent 20:35
Layla_91 flussence: it is ok if I retry manually as long as it continues where it was disconnected.. I am reading the man page.. I still not feel so sure the md5 of the file will be correct after 3-4 disconnections :D
20:36 mtk left
flussence it's TCP, so it *should* be safe from corruption in theory... 20:36
(crc32 every 1.5KB or so, iirc)
jnthn back
benabik flussence: curl does have a --retries option, but I'm not 100% sure it plays nice with -C - 20:37
Layla_91 flussence: TCP is great... but it is wget that might not be perfect :)
jnthn: Prevet! ^_^
moritz_ Layla_91: in my experience wget does very well.
benabik flussence: Sorry, --retry (details)
moritz_ it's a mature project 20:38
benabik Layla_91: I've never had a problem with curl or wget saving bad data when interrupted.
20:38 mtk joined
flussence benabik: I know, but I checked the manpage and it doesn't have an equivalent to wget's 0 == Inf 20:38
benabik flussence:
Layla_91 moritz_: I will give it a try then =) 20:39
benabik flussence: True... I like 0 meaning 0 myself. ;-) But you can set it to something silly like --retry 10000. If 10k retries don't work, then you might want to figure out why. :-D
flussence heh :)
jnthn Layla_91: Привет! Как дела? Я в Стокгольме! :-)
jnthn likes being in a bigger city for a week. :-) 20:40
Layla_91 jnthn: Lucky you! =) 20:42
lichtkind jnthn: <klonk> <- sound of an opening beer bottle :)
jnthn Layla_91: Yeah, but I'm teaching all day long. So I only get to see the city by night.
lichtkind: Nah, I have to be disciplined this week. :P 20:43
Layla_91 flussence: so should I use: wget -c -tries=0 --waitretry=5 $url;
jnthn The evenings are for preparing the next day's class, not beer. :(
lichtkind jnthn: what is model6 is doing?
jnthn lichtkind: Do you mean, what is it for? Or how is it going? 20:44
lichtkind: If the first, I wrote a doc to try and explain that: github.com/jnthn/6model/blob/maste...erview.pod :-) 20:45
lichtkind the ladder, some minimal investigation from my side you can presume
jnthn lichtkind: Going well overall. Next week I start to get Rakudo to use it. Will be in a branch.
Layla_91 jnthn: mmm.. teaching is a little hard I think O_o.. but still .. stockholm! :D I do not like beer a lot.. I prefer eating Chocolates! :P 20:46
Tene lichtkind: 6guts.wordpress.com/ is a good place to follow jnthn's progress
if you want more detail than that, the commit log is pretty enlightening
jnthn Layla_91: Yes, it's very tiring. But my students are lovely. :) 20:47
Layla_91: You'd love the place I'm teaching. There's lots of chocolates around. ;-)
lichtkind Tene: thanks a lot 20:48
Tene lichtkind: thank jnthn++
Layla_91 jnthn: hihi! =)
flussence Layla_91: er, late answer, but yes. with s/-tries/--tries/ of course. 20:49
20:51 orafu joined
Layla_91 flussence: okeys.. here we start.. it is a 3.5GB iso image.. and here it starts! :) 20:51
20:52 Mowah left
Tene Layla_91: alternatively, you could look for a different source, like rsync. 20:52
if it's a distro install image or something, you can often find an rsync server 20:53
Good luck, though.
Intermittent internet connections are a pain.
lichtkind Tene: "jnthn is great" is a consant string :)
Layla_91 Tene: yes they are :\ 20:54
20:54 kaydsoft_ joined
Layla_91 Tene: I just found an app called fatrat.. funny name but seems ok too 20:54
20:55 kaydsoft left 20:56 plainhao left 20:59 mberends joined, kaydsoft_ left, kaydsoft joined 21:05 bluescreen left, benabik is now known as benabik_school, benabik_school left, benabik_school joined 21:06 benabik_school left 21:14 Chat9170 joined
Chat9170 hi guday ppl wats up 21:15
masak hi Chat9170!
Perl 6 is up! and running! :)
21:16 Chat9170 is now known as loveaboy
mberends and chatting :) (hi masak :) 21:16
tadzik oh hi mberends :)
moritz_ \o
loveaboy scn
masak g'day mberends. \o
jnthn mberends! o/ 21:17
mberends hi moritz_ & tadzik & jnthn !
21:17 bluescreen joined
loveaboy brb 21:17
masak the 'Net is fantastic. for the first time in the history of mankind, we can self-select communities.
(rather than finding us in one, geographically)
21:19 bluescreen left
Layla_91 masak: that is bad too.. people became too selective instead of being adaptive.. (or maybe it is good? :S) 21:19
21:19 bluescreen joined 21:21 loveaboy left
masak Layla_91: a little bit of both, I guess. the option to leave when people are being destructive is easier when geography is not involved. 21:21
and so the people who stick around are likely to be the ones who care. 21:22
moritz_ if I were to only meet the p6 hackers in 50km distance or less, I'd be pretty lonely with my hacking hobby 21:23
mberends internetters need to be more selective now that everyone can shout with global scale megaphones.
Tene masak: Have you seen Eliezer's "Evaporative cooling of online communities" extended metaphor?
masak Tene: no, haven't dug into Eliezer beyond MoR yet. 21:24
Tene masak: lesswrong.com/lw/lr/evaporative_coo...p_beliefs/ 21:25
sjohnson sweet link 21:26
Tene "It's the articulate trolls that you should be wary of ejecting, on this theory - they serve the hidden function of legitimizing less extreme disagreements."
masak interesting. 21:27
Layla_91 masak: hopefully someday even seeing P6 hackers geographically will be as easy too =) (for example.. I see java guys where ever I turn.. :| )
Tene See also blog.bumblebeelabs.com/social-softw...ng-effect/
masak ETOOMUCHWEB 21:28
21:28 y3llow_ joined
moritz_ waits for MoR chapters 21:28
21:28 pothos_ joined
masak same. 21:28
21:29 y3llow_ is now known as y3llow
Tene There was a luminosity update today. 21:29
It's not as good as MoR, but updates more-frequently.
21:30 MayDaniel left 21:31 pothos_ is now known as pothos
moritz_ whenever I say "oh, there's another chapter of MoR online" at home, my wife says "you aren't going to read the whole thing today, are you?", then takes a look at me and a resigned breath 21:35
fortunately I'm not too slow at reading
masak MoR: threat or menace to Perl 6 coders' tuits? :) 21:36
PerlJam dare I ask what MoR is? 21:40
masak PerlJam: I tend to spread the longer ones out a bit over several sessions. I read
oops. 21:41
PerlJam: www.fanfiction.net/s/5782108/1/Harr...ationality
PerlJam oh!
moritz_ PerlJam: DON'T READ IT
masak it's all the rage on #perl6 lately. mainly because moritz_ and I like it and won't stop recommending it :)
PerlJam I've tried to steer clear of that as I've already got enough distractions
moritz_ you might not be able to stop it
masak PerlJam: let us assure you, it is a distraction. 21:42
PerlJam have you seen ninjerktsu.blogspot.com/2011/01/car...armed.html ?
21:43 mberends left 21:44 pmurias joined
tadzik you're tempting me to read it 21:44
Tene tadzik: It's the best fiction I've read recently, fwiw. 21:45
pmurias ruoso: re coroutines in javascript i used www.neilmix.com/narrativejs/doc/ once (the tool itself is bad so don't use that) 21:46
21:47 kfo_ joined 21:51 kfo left, Trashlord left
Layla_91 good night all o/ 21:52
diakopter o\
er 21:53
./
er
O/
coldhead /o\
sjohnson heh
Layla_91 diakopter: your special dance? :P 21:55
sjohnson cute emoticons
21:55 Trashlord joined 21:56 mberends joined
Layla_91 I like this one »ö« 21:56
21:56 stkowski joined 21:58 mberends left
sjohnson （　｀ー´） 21:59
Layla_91 chinese one? ('_') 22:02
jnthn Layla_91: good night o/
sjohnson heh 22:03
Layla_91: my version was the fatface version
Layla_91 jnthn: spokoynaya noch! (will buy russian keyboard on may :) ) 22:04
jnthn :) :)
22:05 Layla_91 left
masak русская клавиатура? Почему? 22:08
imagine having a keyboard for every language you needed to write in... :)
s/language/script/ 22:09
tadzik a Perl keyboard would be nice
Tene optimus keyboard? 22:11
www.artlebedev.com/everything/optimus/
22:12 uvtc left
masak the Optimus keyboard is a nice gadget, I guess. but it's nothing I feel a great need for, since I touch-type. 22:13
jnthn masak: Says you with the 4000 char Chinese keyboard taking up half your appartment :P
masak man, I hate that keyboard.
it blocks half my bed, too. 22:14
some days I wake up with the characters 洋葱 imprinted on my forehead. 22:15
also, "reaching for the mouse" means getting up and going into the kitchen... 22:16
jnthn At least the Perl foundation would like your forehead... :) 22:17
masak I might do the last of Web.pm on such a morning... :P 22:18
22:29 stkowski left 22:34 kjeldahl left
masak decides to do some more Yapsi refactoring before going to bed 22:46
last night before I switched off, a bunch of tests started passing again. :) 22:47
22:48 ab5tract joined
lichtkind masak++ 22:54
colomon \o/
colomon is getting Math::FatRat warmed up.
lichtkind holds his tongue that no cheap jokes with punks and fat rats can slip 22:56
23:00 kaare_ left
colomon rakudo: my $a = 10; say q<this is a $a test> 23:02
p6eval rakudo 2666b6: OUTPUT«this is a $a test␤»
colomon rakudo: my $a = 10; say qq<this is a $a test>
p6eval rakudo 2666b6: OUTPUT«this is a 10 test␤»
PerlJam rakudo; my $a = 10; say <<This is a $a test>> 23:14
rakudo: my $a = 10; say <<This is a $a test>> 23:15
p6eval rakudo 2666b6: OUTPUT«Thisisa$atest␤»
masak known bug. 23:16
flussence rakudo: my $a = 10; say «This is a $a test» 23:21
p6eval rakudo 2666b6: OUTPUT«===SORRY!===␤Can't form :w list from non-constant strings (yet) at line 22, near ""␤»
flussence why the difference?
masak good question. there shouldn't be one. 23:23
23:27 whiteknight joined
Tene en.wikipedia.org/wiki/Haecceity -- Intersting terms. 23:36
masak 'night, #perl6
23:37 felliott left, masak left 23:43 ab5tract left 23:51 pmurias left

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!